68 Nielsen

8/13/2019 68 Nielsen

1/8

Session 15: Effects of Moisture 2 Building Physics 2002 - 6th Nordic Symposium

635

Use of FMEA - failure modes effects analysis

on moisture problems in buildings

Anker Nielsen, Professor Ph.D.

1. INTRODUCTIONThe traditional method used for solving moisture failures, as a water leakage in a building is

to make an investigation and find the cause and repair the leakage. This solves the problem

now, but it does not always prevent future failures. We need a systematic method to avoid

future moisture problems. Solving similar failure problems is important in many industries as

in production of aircrafts, cars and computers. The producers want to prevent failures by

analysing the systems before production and getting a feedback from accidents. This is well

known in the airline industry, where every accident is analysed to find the cause and after

each accident a number of changes in the design or in the maintenance is suggested.

The building sector is different from most stationary industry as only part of the building pro-

cess is industrialized and the rest is done as craft. Many operations are not repeated and the

work is adjusted on site. The workmanship is therefore important for the result. An analysis

must include both technical and human errors.

We are interested in moisture problems in buildings, as we know that living in a moist buil-

ding give a higher risk for health problems (Bornehaug et al. 2001). We do not know if it is

emissions or mould or other factors that is problem. We have to look at too high moisture

content (condensation and free water) as a problem. Finding methods to reduce the moisture

failure modes will be economic both from a maintenance and indoor climate perspective.

2. MOISTURE CHECK LISTSToday we try to prevent many of the moisture problems by using check lists with the points

you have to think about, when you design a building. An example is found in Samuelson and

Nielsen 2002. These lists have a tendency to be rather long and include both minor and major

problems. There is no information of the consequence of not following the rules. Having a

better understanding of the risk involved with the failures would make it more evident, when

to use extra control and/or make a change in design to reduce costs of future failures.

3. FAILURE MODE AND EFFECT ANALYSIS (FMEA)

What is a failure? Failure is defined as the inability of any asset to do what its users want it todo. A failure mode and effect analysis (FMEA) as described in US MIL STD 1629 (1980) and

Moubray (1999) are intended to recognize and evaluate the potential failures of a product (the

house) or process (construction) and find its effects. It will identify actions that can be taken

to prevent failures by eliminating or reducing the risk. The concept of FMEA is nothing new.

Designers have always thought of failure modes and how to prevent them. In construction we

have many rules and guides of how to build a house without future failures, but in most cases

we do not look at the effect of the failure. The FMEA method uses a more systematic

approach.

SP Swedish National Testing and Research Institute, Energy Technology, Building

Physics, Box 857, SE-501 15 Bors, Sweden

8/13/2019 68 Nielsen

2/8


636

There are two types of analysis. The first type is the design FMEA, where an evaluation is

done at the design stage. At this stage it can be very helpful to identify possible failures and

establish priorities based on severity and uncover oversight, misjudgements and errors that

has been made. In the process many failures can be prevented before the house is build and

also define where it is necessary to test or control the solutions. The second type is the as

build FMEA where we look at a finished building and identify possible failure modes. Thisis in other applications called Process FMEA as most FMEA is made on production pro-

cesses in the industry.

4. THE BUILDING AS A SYSTEMThe FMEA will typically begin with a number of block diagrams. The purpose is to under-

stand the logic in the system. In traditional building physics we look at moisture problems in

constructions and do not look at the water and heating installation. But to prevent damage

from water systems it is important to go to the source and also look at the installations.

Roof Walls

Floor Space

Room

Water system Heating system

Installation

Rain and snow Ground water

Flooding Internal moisture

Temperature

Climate

Building

Figure 1. The building as a system

Figure 1 views the building as a system divided in three parts: the room, the installation and

the climate. The room is divided in the different structural parts as roofs, walls, floors and the

space (humid air). The installations are divided in water system for cold and warm water and

the water based heating system. The climate includes the effect of rain and snow, flooding

and ground water and internal moisture sources. The next phase is to subdivide the technical

systems as for instance the water installation in figure 2.

Metal

Plastic

Pipes

WC Hand-wash basin

Bath Pump

Washing machine Tank

Dish-washer Water heater

Equipment

Showers

Taps

Bend

Joints

Fittings/Joints

Water installation

Figure 2. Water installation

On the top level we divide it in pipes, fittings, taps and equipment. Pipes are single length

pipes without joints and fittings. All joints between pipes and between pipes and fittings or

equipment are taken in a single group, as these are weak points with a higher risk. Taps are

8/13/2019 68 Nielsen

3/8


637

points with a high risk if left open. Showers are in the same group. The equipment group con-

sists of all the different types of water-connected equipment.

Analysing a water installation as a separate system can be done in a FMEA analysis similar to

many process systems in the industry. But the analysis cannot be done without information on

the building, as a leakage will have different consequences depending on where it occurs. Inrooms with drain the water can run out without making damage. The repair costs will be low.

In rooms without drain as living rooms or worse in hidden spaces will a leakage give a more

extensive damage and the repair cost can by high if a leakage is not stopped immediately.

We also need to analyse the room in the building in more detail. We can make a diagram as

for the water installation that describes the parts as walls, roofs, floors and spaces. This can be

further divided. An example is that the wall is divided into three parts the outer, the inner

surface and the central part. On the inner surface we can get running water or condensation

in both cases we are able to see it and make repairs. Inside the structure we can get interstitial

condensation and high moisture content from water or condensation. The outer surface is only

interesting, if the water moves into the structure. The difficulty in cases with water inside thestructure is that moisture problems are hidden until we see it on the inner surface or we smell

mould in the room. At that time the repair cost could be rather high.

The spaces or rooms can be divided in wet rooms and dry rooms. Wet rooms as bath, washing

and boiler rooms have a high risk for water leakage and therefore must have a drain. Dry

rooms as living rooms, sleeping rooms, kitchen, entrance halls and storage rooms are made

without drain. A water leakage in these rooms can be expensive. Note that the kitchen is

placed in this group, as there is normally no drain from the floor. But it is normal to make

kitchen floors so that they are water resistant but not always watertight. Water drops or water

spills can then happen without making damage.

5. FAILURE TYPESPotential failure modes in industry are modes where the system fails totally or not being able

to output the expected amount or quality. Moisture failure modes will include excess moisture

in the building. Main failure modes in the top (1.level) can be divided in:

Free water in the building - example leakage from water installations Surface condensation - example bathroom after bathing Internal condensation in the structure roof without a vapour barrier

Table 1 shows part of a failure mode table for moisture problems in buildings. It includesfailure modes from free water in the building (level 1). In the next level (level 2) we describe

the causes that can give a level 1 problem. Two causes is described in more details rain water

leakage and leakage from the drainage system. It is often thought that we can find root

causes in the analysis. It implies that we can find a final and absolute level of causation. This

is very seldom the case. We will generally be able to go in still more detail with the causes. In

the table 1 we have stopped at level 5. The information is based on (Andersson and Kling

2000), (Nevander and Elmarson 1994), (Nielsen 2000) and (Norwegian Building Research

Institute 2001). The total table of failures modes is 6 pages.

6. FAILURE EFFECTS

After the failure mode analysis we must find the failure effects - What happens when a fai-lure mode occurs. The failure effect describe if obvious physical effects as unusual smell,

8/13/2019 68 Nielsen

4/8


638

visible moist spots or pools of water on the floor accompany the failure. Table 2 is a table

with description of failure effects for some failure modes in table 1. After making a list of fai-

lure effects it is possible to write prevention methods for each failure as in the table. This

could for instance be to make tests and check of the workmanship quality.

Table 1.Failure modes from liquid water in the building (level 1).Level 2 Level 3 Level 4 Level 5

Leakage rain water Leakage from roofs Mechanical damage

Montage error

Leakage in roofing

material

Aging of material

Leakage from

skylights

Montage error

Mechanical damage

Leakage from roofing

felt joints

Aging of material

Leakage from guttersLeakage from joint to

walls

Montage error

Design error

Leakage from joints

with pipes, cables and

duct going through the

roofCracking from

movements

Leakage from drain

system

Leakage from pipes Mechanical damage

Corrosion

Too high temperatures

Leakage from pipe

material

Constant water

pressure

Leakage from joints

between pipes

Montage error

Leakage from fitting Montage error

Montage errorLeakage from drains in

floors Filled with solid waste

Leakage trough

material

Leakage trough joints

Leakage joints with

walls

Leakage from

watertight floors (ex.

bathrooms)

Leakage joint with

pipes, cables and ducts

The same failure effects from moisture in buildings can have different causes. An example is

condensation seen on a surface; this could come from a thermal bridge, a low ventilation rate,

and extra moisture in the indoor air from cooking, bathing and so on. Another example is

failure effects from free water in the form of dripping or water flows. This could come from

leakage in the water or drainage system, leakage in the heating system and drips fromcondensation on the surface or in the construction. In some cases is the source obvious as a

8/13/2019 68 Nielsen

5/8


639

leakage from a dishwasher. In other cases is the source hidden and it is not possible to find the

failure mode without having drawings and/or opening of the construction.

Table 2.Failure modes, failure effects and prevention methods for leakage in roofs.

Failure modes Failure effect Prevention method

Leakage inroofing

material

Mechanicaldamage

Seen as dripping or water flowin the building during and after

rain. Location of damage can

be very difficult.

No normal traffic on theroofing material. Remove all

waste material on roofs. Inspect

roof.

Montage error The roof leaks shortly after

construction. Seen as dripping

or water flow in the building

during and after rain. Location

of damage can be very difficult.

Control system for

workmanship. The contactor

must repair the damage.

Aging of

material

The material cracks from frost

damage or temperaturevariation.

Use materials tested for aging

in a climate as on site. Inspectroof and look for cracks or

other damages

Leakage from

skylights

Seen as dripping or water

flowing in the building at the

skylights during and after rain.

Use skylights that have been

tested for water tightness. Have

control system for workman-

ship.

Leakage from

roofing felt

oints






Control system for

workmanship. The contactor

must repair the damage.

Mechanical

damage

See mechanical damage on

roofs

Aging joints See aging of roof materials

Leakage from

gutters

Water damage on facade from

gutter leakage. Possible water

damage into constructions.

Control gutters for leakage. In

case of repeated leakage should

gutters be repaired.

Leakage from

oint to walls

The roof leaks. Seen as

dripping or water flow in the

building during and after rain.

Location of damage can bevery difficult.

Roofing material should

continue approx 10 cm up

connecting walls. Rain should

not be able to come under theroof material at the joint.

Leakage from

oints with

pipes, cables

and duct in

the roof






Montage of signboards or

ventilation equipment on the

roof must be specially checked

to ensure watertight solutions

Design error As above Areas around pipes, cables and

duct must be planned, so it is

easy to make watertight

solutions.Cracking/

Movements

As above Look for cracking around

pipes, cables and ducts.

8/13/2019 68 Nielsen

6/8


640

2. FAILURE PATTERNS AND RISK EVALUATIONWe have an intuitive expectation, that the failure rate is low for new things and it is will in-

crease with time. This is incorrect in many practical cases. The classical failure pattern is

called the bathtub from the drawing of the risk. There is a high-risk of for instance a water

leakage in the start and the end. In between the risk is lower. The infant mortality in the startis caused by: poor design, poor quality manufacture, incorrect installation, incorrect opera-

tion, unnecessary maintenance and bad workmanship. We know that infant mortality is a pro-

blem for many systems as many parts are made or installed on site. An example is water

based heating systems. They have a high risk in the start from incorrect installation. Then the

risk goes down. In the end it will increase from wear out from for instance corrosion. Other

systems will have no infant mortality but a failure risk that will increase with time.

The FMEA analysis can be expanded with a risk evaluation looking at probability for loss of

function, but this is difficult for buildings, as we do not collect systematic information on the

failure risks from different constructions. Calculation of risk can sometimes be done as for

condensation in a wall (Nielsen 1995). In most cases we can intuitively state that certain solu-tions are more risky than another as placing a water tap in a room without drain. A compli-

cation is that the many failure effects are time-dependent.

Information from the Norwegian Building Research Institute (2001) shows, that around 80%

of all investigated building failures is related to water and moisture. 70% of the moisture fai-

lures are related to water leakage and moisture transport, 20% is condensation and 10% is

moisture from the building phase. The Danish Building Research Institute (Valbjrn and Erik-

sen 2001) has made an enquiry to professionals in the building sector about the risk for moi-

sture damage in buildings. The participants were asked to mark the risk level as a point on a

line going from little risk to high risk for each type of construction. Some of the results were:

Installations presented a comparatively great risk of water damage. Pitched roofs without dormers or valleys run less risk of water damage than flat roofs

or other types of roof with dormers and valleys.

3. USE OF FMEA ON ACTUAL BUILDINGSThe description until this point is in general terms. For practical use of the method we must

make the analysis on an actual buildings. This is done with the following procedure:

Get the drawings of the building, the drawings of the water, heating and wastewater instal-

lation and the description of the solutions. In an existing building an inspection (make notesand take pictures) is necessary to check if the drawings describe the as build case.

Traditionally in building physics we look at the different constructions as floors, walls and

roofs. In this case we are also interested in getting a good indoor environment without moi-

sture problems and will start looking at the rooms.

Make a list of all possible moisture sources in each room. First include internal water

installation, water-heating installation, drain system, periods with high moisture content in air(from bathing, cooking and so on). Second include moisture from rain and snow, surface

water and ground water.

Then start by looking at the water installation where is high-risk points, taps, fitting and soon. Each point is giving a risk value from 0 to 10. These values are then reduced if the room

8/13/2019 68 Nielsen

7/8


641

has a drain and a watertight floor or other risk reducing methods is used. An example is a disk

washing machine that will have a risk of 10. If we place it on a watertight floor with drain the

risk is reduced to 3. The risk can still be reduced, if we always close the water tap after each

used or we use a break preventer valve, which will stop the water flow if the tube between the

tap and the machine breaks. A list of risk values will be found in the final description of the

method. The result is risk values for each room and a value for the whole building. Similarmethods must be used for all other moisture sources. In looking on the possible causes we use

our FMEA analysis as a basis and supplement with cases that is not covered in the general

list. Make a total risk value for each room in the building. The risk values give the possibility

to compare different solutions.

In the design phase we can specially look at rooms with high risk values and try to reduce the

risk values by changing the construction and installation. Some of these reductions can also

come from a better quality control of the work. In that case it is important that this informa-

tion is given the workmen at the site. For the finished building we can see where we have risk

points, and that is where we should check in the operation phase.

The method described here can also be done existing buildings. The analysis of an existing

building should be done with the people responsible for the maintenance as they known

previous moisture problems in the building. The result of an analysis of an existing building is

to help in the maintenance and give a better understanding of where future problem could

come.

A problem with the method is to give risk values 0-10 to different failure types and reductions

for prevention methods. Making the list should be done by a group of people with experience

from practical moisture problems including maintenance people.

A more comprehensive report with examples of the methods use will come from the project.

4. CONCLUSIONSA FMEA analysis is a good help in finding better solution for moisture proof buildings. Using

this systematic approach gives better understanding of building failures, their effects and re-

mediation methods. Finding and preventing hidden failures is a very important task. Using the

right solution in the constructions can also reduce the risk of serious damage from water lea-

kage. The analysis is important in the building phase but must include the influence of the

user and the lifetime of the building.

The analysis is also very useful in facility management (FM). The building owner will havean interest in the keeping the running cost down and still get a good indoor climate. The

management of the building can collect very much information on the cost and types of fai-

lures during the life cycle.

The user can increase the risk of moisture problems in the building. The behaviour can give a

much higher risk of problems without the user knows that this is the case. Making information

for the user is a very good idea that can save money in the management of the building.

FMEA analysis on buildings is a method for better quality of the buildings, as results from re-

search and practise is combined. The analysis results as checklists and information on critical

points should be structured for the different parties in the building process as architects, engi-

8/13/2019 68 Nielsen

8/8


642

neers and craftsmen. Selecting, building and keeping moisture-proof constructions are impor-

tant for preventing health problems in buildings.

5. ACKNOWLEDGEMENTSThe work is supported by FORMAS- the Swedish Research Council for Environment, Agri-

cultural Sciences and Spatial Planning and the International Centre for Indoor Environmentand Energy at the Technical University of Denmark6. LITERATUREAndersson, J. and Kling, R. 2000: Bygg Vattenskadeskert VASKA viser vgen, (Build

water-proof VASKA project) Swedish Building Research Council, Sweden, report T3 2000

Bornehaug, C. G. et al. 2001: Dampness in Buildings and Health. Nordic Interdisciplinary Re-

view of the Scientific Evidence on Associations between Exposure to Dampness in Buil-

dings and Health Effects (NORDDAMP), Indoor Air, vol. 11, no 2, page 72-86

Moubray, J. 1999: Reliability-centered Maintenance, 2.edition, Butterworth-Heinemann, Ox-ford, UK, ISBN 0 7506 3358 1

Nevander, L. E. and Elmarson, B. 1994: Fukthandbok (Moisture Handbook), 2nd edition,

Svensk Byggtjnst, Stockholm, Sweden

Nielsen, A. 1995: Use of statistics for prediction of condensation in a wall construction. Inter-

national symposium Indoor Air Quality in Practice, Moisture and Cold Climate Solutions,

ISIAQ, Oslo, Norway, 19-21 June 1995, page 206-216

Nielsen, A. 2000: Analysis of the moisture problems in bath rooms, Healthy Buildings 2000,

Helsinki, Finland, August 5-10, vol. 3, page 495-500

Norwegian Building Research Institute 2001, Building Detail Sheets, Oslo, Norway, new

update every year

Samuelson, I and Nielsen, A. 2002: Checklists for clean and moisture safe buildings, 6th Con-

ference on Building Physics in the Nordic Countries, Trondheim, Norway

US MIL STD 1629 1980: Procedure for performing a Failure Mode, Effects and Criticality

Analysis, method 102, 24 November 1980

Valbjrn, O. and Eriksen, S. 2001: Bygningskonstruktioners risiko for fugtskader, Erfaringer

fra praksis, (The risk of moisture damage in building constructions, experience from practice),

By og Byg Resultater 012, Danish Building Research Institute, Hrsholm, Denmark

68 Nielsen

Documents

Transcript of 68 Nielsen